As someone who looks at technology stocks on a daily basis, overvalued growth stocks aren't anything new to me. However, over the last several months, I've singled out one specific company repeatedly as a company that I believed could be headed for a pullbackPandora Media (P) for a number of reasons.

Perhaps the most high-profile reason was the recent introduction of Apple's (AAPL 0.51%) iTunes Radio, which is essentially a Pandora clone. I wasn't alone in thinking Apple could disrupt Pandora, whose stock sold off more than 10% when Apple recently announced that more than 11 million users had already tuned into iTunes Radio in less than a week since its global launch.

Apple expects to surpass 700 million total shipments of iOS devices any day. Rest assured, between Apple's already-massive installed base and its deep pockets, that Apple can and will seriously compete with Pandora for listeners. And although Pandora's reach does extend well beyond Apple devices, a serious competitor like Apple isn't to be underestimated. Ever.

However, my knocks against Pandora were based on another idea, and after digging much, much deeper into Pandora's inner workings over the weekend, I've changed my mind. Turns out, I was, for lack of a better word, wrong, and I'm here today to own up it publicly.

Opening Pandora's box
Now, I need to begin this with a caveat. In saying I was wrong about Pandora, I'm only addressing one specific (albeit probably the most important) aspect in my argument against the company. The cornerstone of my callouts was the idea that Pandora's model simply didn't scale well, which seems pretty understandable when you look at Pandora's financial performance over the last several years.

Fiscal Year*

2009

2010

2011

2012

2013

Revenue ($ million)

$19.3

$55.2

$137.8

$274.3

$427.1

Operating Income

($27.4)

($15.5)

($0.3)

($11.0)

($37.7)

Net Income

($28.2)

($16.8)

($1.8)

($16.1)

($38.1)

Source: Capital IQ; * denotes Pandora's fiscal year used to end in January. For example the 2009 period above reflects from 2/1/08 to 1/31/09.

Looking at the above, it's easy to see how someone might be fooled into thinking Pandora's current model doesn't scale. The company's losses in FY 2013 were actually worse than its losses five years earlier, despite Pandora's revenue having increased more than 2,000% in the interim.

Not exactly encouraging, right? So what did I miss?

Digging deeper
One of the problems of looking at the past to predict the future in investing is that businesses don't operate in a vacuum. They react, adapt, and ideally improve over time, and Pandora has been no exception.

One trend that's sweeping the entertainment or media space is the rising cost of content. Unfortunately, Pandora is no exception there, either -- with the fixed rates it pays its licensing partners still set to tick steadily upward over the next few years.

Year

Cost Per Song for Non-Subscription Listener

Cost Per Song for a Subscription Listener

2012

$0.0011

$0.0020

2013

$0.0012

$0.0022

2014

$0.0013

$0.0023

2015

$0.0014

$0.0025

Source: Pandora annual report. 

As a company with no history of profitability, it's fair to think that rising content costs would keep that trend intact for Pandora, which it turns out just isn't the case.

Returning to the whole "not operating in a vacuum" idea, Pandora has made more substantial strides in making money off each user than I'd given it credit for.

Pandora's growth has been driven in part by an explosion in its user base, which has increased from 3 million monthly active users (MAUs) at the start of 2009 to well more than 71 million at the end of its most recently quarter. But more importantly, Pandora has also significantly improved the amount of revenue it generates from each user as well over the last several quarters.

Metrics that matter most
Perhaps the most crucial financial metric that Pandora tracks are RPMs, or revenue per milli, which is ad-speak for revenue for every 1,000 advertisements served. These have improved dramatically over the last year as Pandora has put significant resources toward beefing up its ad sales teams and technologies that better help the company target specific demographics.

In its most recently reported period, Pandora saw its overall non-GAAP RPM increase 30% from $31.90 in the same quarter last year to $41.73, meaningfully outpacing the increase in content costs highlighted above. With per-play costs largely fixed, this huge improvement in the amount of money generated with each ad helped improve Pandora's gross margins by 900 basis points year over year from 34% to 43% last quarter.

This is exactly the changing dynamic that should help shift Pandora toward eventually generating sustained profits. For its current fiscal year, the average analyst expects Pandora to earn $0.03. However, with the improved operating leverage from higher RPMs flowing through to the bottom line, Pandora is expect to produce $0.26 in per share earnings next year and never really look back.

So the problem here isn't that Pandora's business model doesn't scale -- it just hasn't hit a point where the benefits of operating at scale have flowed all the way through to the bottom line yet.

That was my big oversight in looking at Pandora.

Not out of the woods yet
This isn't to say I've drunk the Kool-Aid when it comes to Pandora.

When I looked at Pandora in the past, I was actually thinking of it as a potential moneymaking opportunity on the short side, so saying I don't believe its business model is fundamentally flawed is still far from a ringing endorsement.

I still think Pandora is still pretty clearly overvalued at nearly 10 times revenue, especially as the online radio and streaming space grows more crowded by the day. Investors are also clearly still edgy about the looming threat of Apple's iTunes Radio. Pandora's stock tanked again this week when Apple announced it plans to roll out iTunes Radio into the U.K. However, only time will tell how great a threat Apple's new and growing presence in this space will present to Pandora.

Foolish bottom line
The real point here is that, in the end, I got this one wrong. Pandora, as it turns out, is a company with a future -- and possibly a very bright one at that.

One of the most important parts of growing as an investor is admitting when you make a mistake. I did just that when I looked at Pandora in the past. I'm sure I'll misjudge another company in the not-too-distant future, but I believe owning up to your mistakes when they rear their ugly heads is one of the most important things a true Fool can do.

My bad, people. My bad.